Multi-objective optimization of a self-propelled capsule for small-bowel endoscopy considering the influence of intestinal environment

Abstract

This work studies the control optimization of a self-propelled capsule moving in the small intestine for endoscopic diagnosis. For this purpose, we combine an existing capsule model with the intestinal peristalsis and its internal environment in order to gain a better understanding of dynamics of the self-propelled capsule. For the optimization study, a number of different realistic targets are considered, including the capsule's average progression speed, the impact force acting on the small intestine and the capsule's energy consumption. In addition, the uncertainty of the small intestine environment is taking into account by varying its internal radius. In this setting, we develop a multi-objective optimization strategy based on NSGA-II, Monte Carlo, and Six-Sigma algorithms considering both the control and structural model parameters, such as excitation frequency and impact stiffness. The effectiveness of the proposed optimization strategy is demonstrated via extensive numerical simulations with the consideration of a wide range of realistic scenarios.